The anatomy of an address
When the internet first became a reality, few could have predicted just how significant it would become in the day-to-day lives of businesses, organisations and the general public. Initial IP addresses were simple 32-bit numbers comprised of four three-digit decimals. Although this formula allows the creation of 4.3 billion addresses, it has become necessary to expand the number and this has resulted in IP V6 coming increasingly to the fore.
When it comes to IP addresses, most installers and integrators will agree that things are relatively simple. With IP V4, the most common form of IP addressing, the 32-bit address is formed of four dot-separated three digit decimals with a value between 0 and 255.
In the vast majority of applications, addresses will be allocated within the same network subnet to connected cameras, encoders and other peripherals. With a minimal amount of record keeping and a planned distribution of addresses, systems can be easily created that are both orderly and logical in terms of how they are addressed.
When the internet was young, 32-bit addresses made sense. No one foresaw the explosion of internet connectivity which is still growing today, and few believed that the internet would catch on, let alone considered the potential of the Internet of Things. Using 32-bit addresses meant there were up to 4.3 billion options. Given that there are also nearly 18 million private IP addresses that can be used on non-internet facing networks (and many security systems make use of these), the possibility of running out of addresses seemed unlikely. However, that’s just what has happened.
Well, in the interests of fairness, we haven’t actually run out of IP V4 addresses, but most of them have been allocated. Many institutions hold blocks of IP V4 addresses that are unused, and as they become more scarce other blocks are being traded. Despite this, a new system is needed.
It must be said that IP V6 isn’t new. It has been around since the late 1990s and started to become more common in the past decade. IP V6 makes use of 128-bit addresses which hugely increases the number of available options. Of course, for those of us who are used to IP V4 addresses, it also vastly increases the complexity of addresses, especially as IP V6 makes use of ‘shorthand’ to remove redundancies in addresses.
Most people will suggest that the sole benefit of IP V6 is expanded address numbers, but there’s actually a bit more to it than that. IP V6 packets are formulated differently, and the headers make routing more efficient as less processing power is required. This is despite IP V6 headers including more information. Routing is more efficient because of the way the header data is formated.
Additionally, IP V6 supports multicast rather than broadcast, as was the case with IP V4. This allows data flows sent to multiple recipients to require less bandwidth.
The reality is that while IP V6 has had something of a ‘slow burn’ in recent years, it is something that installers and integrators should be getting used to.
Concerns for installers?
A few years ago much was made of the fact that IP V4 and IP V6 used different protocols and so weren’t interchangeable. Today most operating systems, switches and other IT equipment, field devices and peripherals are capable of supporting both IP V4 and V6.
For the vast majority of security applications, static IP addresses for private networks are used. There are some in the industry – typically those providing field devices with limited or no installation utilities – that default devices to DHCP because it makes their job simpler. However, inevitably installers and integrators will disable DHCP and use static addressing.
The reason for this is simple. If a critical device is on the security network, it is important that its IP address is always known and always the same. Because of the always-on nature of security systems, there is no benefit in having to look up addresses or use LAN scanning tools when a specific device needs attention.
Indeed, recently during a field trial of an advanced system, one technical engineer with a very good knowledge of IP-based networking commented, ‘There is no place for DHCP in security; I need to know where any device is at any given time with 100 per cent certainty’.
At a LAN level, most security systems make use of private IP addresses. In IP V4 these addresses are 192.168.0.0–192.168.255.255; 172.16.0.0– 172.31.255.255 or 10.0.0.0–10.255.255.255. Private IP addresses can be replicated across a wide number of sites because they are non-internet facing. In other words, they cannot be connected to via the internet. Instead the router has two IP addresses: one is internet facing and other is private. The connection between the two is achieved with port forwarding.
IP V6 will also enable static addresses and will have certain addresses reserved for non-internet facing networks, so to all intents and purposes there are no concerns when it comes to creating security systems; things will remain much the same in terms of network structure.
When it comes to creating addresses, working with IP V4 is simple and logical. However, IP V6 does present challenges and for those approaching it for the first time it can be complex and challenging.
IP V6 addresses are made up of eight groups of four hexidecimal digits (0–9 and A–F) separated by colons. There are four types of addresses: unicast, multicast, anycast and link-local. These have slightly different configurations.
Where a group or groups solely consist of zeros, these can be replaced with two colons. However, this can only happen once in the address. Also, leading zeros can be omitted but each group requires at least one value. By way of an example, consider the following IP V6 address.
This could be written as follows: 2001:604::b3ff:0:0:329, or 2001:604:0:0:b3ff::329, but not as 2001:604::b3ff::329!
Asking people how long it will be before IP V4 is obsolete is akin to asking how long is a piece of string. Responses vary from ‘within 18 months, as I’ve been saying since 2005’ through to ‘not in our career lifetimes’.
As with most transitional technologies, the real answer is that the transition will be complete when manufacturers no longer offer IP V4 support in their products, and given the low levels of IP V6 adoption today, that won’t be for a while yet.
In the immediate future, little will change, but it’s best to at least get an understanding of IP V6 and how it will impact on working practices. After all, in wider technological fields technology moves fast, and it’s best not to be caught out!